Foundational Papers in Complexity Science pp. 825–865
DOI: 10.37911/9781947864535.26
Portrait of the Artist as a Young Complexity Hacker
Author: Simon DeDeo, Carnegie Mellon University and Santa Fe Institute
Excerpt
Sometimes good things come from taking mathematics seriously. Artificial intelligence, for example, received a jump-start when Steve “Slug” Russell, a twenty-something hacker at MIT in the 1960s (Levy 1984), took a mathematical function seriously enough that he made it run on an IBM 704; it was the first interpreter for the computer language LISP (McCarthy 1981), and the ultimate ancestor to the Python code that runs our scientific and engineering projects today. Computer science often lives on this knife’s-edge between mathematical order and fuse-blowing reality—the thrill of making mathematics run is part of the attraction of a discipline that picks up a virtual concept and asks “what would happen if this were a thing?” Antics like Russell’s were one reason that nobody knew where to put computer science departments; some came out of electrical engineering, others applied mathematics, meeting in a no-man’s land that even today makes the more disciplinarily inclined nervous.
Around the same time as Russell was putting one mathematical ghost into the machine, an even younger hacker, Gregory Chaitin, was working on something even stranger. The object Chaitin picked up was the ur-model of computation itself—the “Turing Machine,” a philosophical thought experiment dreamed up by Alan Turing (1937) in Cambridge years earlier. In 1965, many would have said this was an unpromising beginning. We had proved, to our satisfaction, that Turing’s model was “complete.” Everything a computer could do, a Turing machine could do. We had gotten past these origins, and had far more appealing, if less philosophical, ways to talk about how to get computers to do what we wanted. These included John McCarthy’s LISP, which Russell was, almost simultaneously, programming on the IBM 704. We had real computers with real problems—ones that made Turing’s mathematical model look positively irrelevant.
Bibliography
Griffiths, T. L., D. Daniels, J. L. Austerweil, and J. B. Tenenbaum. 2018. “Subjective Randomness as Statistical Inference.” Cognitive Psychology 103:85–109. https://doi.org/10.1016/j.cogpsych.2018.02.003.
Levy, S. 1984. Hackers: Heroes of the Computer Revolution. Garden City, NY: Doubleday.
Liu, Y., and R. Pass. 2020. “On One-way Functions and Kolmogorov Complexity.” In 2020 IEEE 61st Annual Symposium on Foundations of Computer Science (FOCS), 1243–1254. https://doi.org/10.1109/FOCS46700.2020.00118.
McCarthy, J. 1960. “Recursive Functions of Symbolic Expressions Their Computation by Machine, Part I.” Communications of the ACM 3 (4): 184–195. https://doi.org/10.1145/367177.367199.
—. 1981. “History of LISP.” In History of Programming Languages, edited by R. Wexelblat. New York, NY: Academic Press. http://jmc.stanford.edu/articles/lisp.html.
Turing, A. M. 1937. “On Computable Numbers, with an Application to the Entscheidungsproblem.” Proceedings of the London Mathematical Society s2-42 (1): 230–265. https://doi.org/10.1112/plms/s2-42.1.230.
Wojtowicz, Z., and S. DeDeo. 2020. “From Probability to Consilience: How Explanatory Values Implement Bayesian Reasoning.” Trends in Cognitive Sciences 24 (12): 981–993. https://doi.org/10.1016/j.tics.2020.09.013.